Sistem Deteksi Marka Jalan Berbasis Convolutional Neural Network

Gustri, Irfan Nanda (2020) Sistem Deteksi Marka Jalan Berbasis Convolutional Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
07211440000017-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (47MB) | Request a copy

Abstract

Keselamatan dalam berkendara sangatlah penting. Akan tetapi pengendara kurang begitu peduli dengan jalanan, salah satunya marka jalan yang bisa mengakibatkan hal yang tidak diinginkan seperti kecelakaan hingga kemacetan lalu lintas. Dengan begitu perlunya pengingat agar pengemudi bisa mengambil tindakan pencegahan, seperti mendeteksi marka yang ada di jalan yang berguna untuk mengontrol dan mempertimbangkan posisi kendaraan. Deteksi marka jalan yang menggunakan Convolutional Neural Network (CNN) sebagai pengindraan kondisi marka jalan sehingga dapat membaca objek badan marka jalan yang mendeteksi dan mengenali bentuk yang ditangkap menggunakan kamera, lalu citra tersebut diterima pada input, diproses, dan output, yang diolah dalam dua dimensi hingga menghasilkan proses yang diharapkan. Library yang digunakan pada bidang CNN ini menggunakan TensorFlow, dibantu dengan algoritma You Only Look Once (YOLO). TensorFlow dan YOLO digunakan untuk mengeksekusi perintah dan mengenali objek yang berbeda. Dengan begitu diharapkan kesalahan yang terjadi di lalu lintas semakin kecil. ========================================================================================================= Safety in driving is very important. But drivers are less concerned with the road, one of which is road markings that can result in unwanted things such as accidents to traffic jam. That way the need for a reminder so that the driver can take precautions, such as detecting markers on the road that are useful for controlling and considering the position of the vehicle. Detection of road markings using Convolutional Neural Network (CNN) as a sensation of road markings conditions so that they can read road mark body objects that detect and recognize shapes captured using a camera, then the image is received on input, processed, and output, processed in two dimensions to produce the expected process. The library used in the CNN field uses TensorFlow, assisted by You Only Look Once (YOLO) algorithm. TensorFlow and YOLO are used to execute commands and recognize different objects. That way it is expected that the errors that occur in traffic will be smaller.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Marka Jalan, Convolutional Neural Network (CNN), TensorFlow, You Only Look Once (YOLO), Road Markings, Convolutional Neural Network (CNN), TensorFlow, You Only Look Once (YOLO).
Subjects: Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
T Technology > TA Engineering (General). Civil engineering (General) > TA1650 Face recognition
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL152.5 Motor vehicles Driving
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User: Irfan Nanda Gustri
Date Deposited: 21 Aug 2020 03:45
Last Modified: 21 Aug 2020 03:45
URI: http://repository.its.ac.id/id/eprint/78934

Actions (login required)

View Item View Item